Dynamic evaluation of wind turbine health condition based on Gaussian mixture model and evidential reasoning

Dong, Yuliang; Fang, Fang; Gu, Yujiong

doi:10.1063/1.4808018

Cited by 20 publications

(11 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…19–21 This has driven researchers and industry toward complex, time-consuming and hardly interpretable models in machine learning such as artificial neural networks (NNs), Gaussian processes and Gaussian mixture models. 22–24 Schlechtingen and Santos, 25 Sun et al, 26 Li et al 27 and Garcia et al 28 propose generalized models for WTs through NNs and fuzzy metrics in order to perform anomaly detection. The procedure shows significant dependence on high quality and large collection of data sets besides the difficulties in defining an accurate threshold technique.…”

Section: Introductionmentioning

confidence: 99%

A health condition model for wind turbine monitoring through neural networks and proportional hazard models

Mazidi

Tjernberg

et al. 2017

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

In this article, a parametric model for health condition monitoring of wind turbines is developed. The study is based on the assumption that a wind turbine’s health condition can be modeled through three features: rotor speed, gearbox temperature and generator winding temperature. At first, three neural network models are created to simulate normal behavior of each feature. Deviation signals are then defined and calculated as accumulated time-series of differences between neural network predictions and actual measurements. These cumulative signals carry health condition–related information. Next, through nonlinear regression technique, the signals are used to produce individual models for considered features, which mathematically have the form of proportional hazard models. Finally, they are combined to construct an overall parametric health condition model which partially represents health condition of the wind turbine. In addition, a dynamic threshold for the model is developed to facilitate and add more insight in performance monitoring aspect. The health condition monitoring of wind turbine model has capability of evaluating real-time and overall health condition of a wind turbine which can also be used with regard to maintenance in electricity generation in electric power systems. The model also has flexibility to overcome current challenges such as scalability and adaptability. The model is verified in illustrating changes in real-time and overall health condition with respect to considered anomalies by testing through actual and artificial data.

show abstract

Section: Introductionmentioning

confidence: 99%

A health condition model for wind turbine monitoring through neural networks and proportional hazard models

Mazidi

Tjernberg

et al. 2017

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

show abstract

“…46 However, as we know, the hidden danger is primarily from the resonance problem between the structure and the excitation from the rotating blades at the frequencies of 1P and 3P. Generally, if the ratio of the difference between the structural natural modal frequencies and the excitation frequencies to the input frequencies is greater than 10%, the resonance effect can be avoided.…”

Section: -26mentioning

confidence: 97%

Operational modal identification of offshore wind turbine structure based on modified stochastic subspace identification method considering harmonic interference

Dong

Lian

Yang

2014

Journal of Renewable and Sustainable Energy

View full text Add to dashboard Cite

The structural modal parameter information can be generally obtained from different vibration responses triggered by unknown excitations using the classic modal identification methods which are based on the assumption that the input to the structure is exactly the same or close to the stationary random white noise. For the actual projects such as the offshore wind turbine structure, however, the imposed loads cannot be considered as the pure white noise excitation because the harmonic components emerge obviously in vibration responses due to periodic rotation excitations of the rotor. When the created harmonic components have greater energy and their frequencies are close to any natural modal frequency of the structure, the classic methods can no longer separate harmonic modes from actual structural operational modes under this strong harmonic disturbance, thus false modes may be generated and the identification accuracy may be badly affected. To identify the actual structural operational modes accurately under strong harmonic excitation, a modified stochastic subspace identification (SSI) method considering harmonic interference called the harmonic modification SSI (HM-SSI) method was proposed in this paper assuming the input harmonic frequencies are known and time-invariant. Then the effectiveness and accuracy of the HM-SSI method were verified through a simple numerical model of cantilever beam excited by various harmonic inputs superimposed on random loads. Besides, the superior robustness and better accuracy of identification results affected by the level of noise and the proportion of harmonic energy were reflected. Finally, the modal parameter information under different operational conditions was obtained and the safety assessment of an operational wind turbine was made based on the measured data from one offshore wind turbine test prototype at high running speed.

show abstract

“…SOM is a variant of the neural network, which can be used to construct the normal behavior model. The minimum quantization error (MQE) is used to assess the state of wind turbine (Dong et al, 2013; Du et al, 2016). The dashed line on right in Figure 10 denotes the moment of the failure, while the dashed line on left denotes the moment when the HI has crossed the threshold drastically.…”

Section: Case Studymentioning

confidence: 99%

“…In recent years, condition monitoring of wind turbine using the SCADA data becomes a popular method (Bangalore et al, 2018; Dao et al, 2018; Schlechtingen and Santos, 2014; Tautz-Weinert and Watson, 2016). For instance, Dong et al (2013) showed a dynamic evaluation of wind turbine health condition method based on Gaussian Mixture Model (GMM) and evidential reasoning (ER). Xiao et al (2014) proposed an operating condition assessment strategy for wind turbine based on fuzzy synthesis method.…”

Section: Introductionmentioning

confidence: 99%

Online health assessment of wind turbine based on operational condition recognition

Zhang

Jiang

et al. 2018

Transactions of the Institute of Measurement and Control

View full text Add to dashboard Cite

To reduce the operation and maintenance (O&M) costs, the health assessment of wind turbine has received more and more attention in recent years. However, it is difficult to evaluate the health condition of wind turbine due to the complex and non-stationary operation environment. This paper proposes a data-driven approach for online health assessment of wind turbine based on operational condition recognition. First, the operational condition parameters are selected by analyzing the monitoring data of wind turbine. Considering the time-varying of operation environment, the operational conditions are divided into four subspaces utilizing the K-means clustering algorithm. Then, using the historical state parameters data under normal operation, a health benchmark model is constructed in each operational condition space based on Gaussian Mixture Model (GMM). Further, a Softmax model is trained according to the results of operational condition classification, which is used to identify the online operational condition of wind turbine. Moreover, an overall health index (HI) based on Mahalanobis distance is developed to assess the health condition of wind turbine. Finally, the method is verified by the actual supervisory control and data acquisition (SCADA) data of a wind field in northwestern China. The test results show that the proposed approach can track the running state of the wind turbine accurately and play a good role in early fault warning.

show abstract

Dynamic evaluation of wind turbine health condition based on Gaussian mixture model and evidential reasoning

Cited by 20 publications

References 22 publications

A health condition model for wind turbine monitoring through neural networks and proportional hazard models

A health condition model for wind turbine monitoring through neural networks and proportional hazard models

Operational modal identification of offshore wind turbine structure based on modified stochastic subspace identification method considering harmonic interference

Online health assessment of wind turbine based on operational condition recognition

Contact Info

Product

Resources

About